Non-specific Adsorption of Crude Cell Lysate on Surface Plasmon

Jul 11, 2013 - Resonance Sensors. Alexandra Aubé,. †. Julien Breault-Turcot,. †. Pierre Chaurand,. †. Joelle N. Pelletier,. †,‡,§ and Jean...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/Langmuir

Non-specific Adsorption of Crude Cell Lysate on Surface Plasmon Resonance Sensors Alexandra Aubé,† Julien Breault-Turcot,† Pierre Chaurand,† Joelle N. Pelletier,†,‡,§ and Jean-François Masson*,†,∥ †

Département de Chimie, and ‡Département de Biochimie, Université de Montréal, C.P. 6128 succursale Centre-ville, Montreal, Quebec H3C 3J7, Canada § PROTEO, The Quebec Network for Research on Protein Structure, Function and Engineering, Université Laval, Quebec City, Quebec G1V 0A6, Canada ∥ Centre for Self-Assembled Chemical Structures (CSACS), McGill University, Montreal, Quebec H3A 2K6, Canada ABSTRACT: Non-specific adsorption of the molecular components of biofluids is ubiquitous in the area of biosensing technologies, severely limiting the use of biosensors in realworld applications. The surface chemistries developed to prevent non-specific adsorption of crude serum are not necessarily suited for sensing in other biosamples. In particular, the diagnostic potential of differential expression of proteins in tissues makes cell lysate attractive for disease diagnostics using solid biopsies. However, crude cell lysate poses a significant challenge for surface chemistries because of a large concentration of highly adherent lipids. Contrary to the nonspecific adsorption in crude serum being suppressed by hydrophilic surfaces, the surface plasmon resonance (SPR) analysis of serine-, aspartic-acid-, histidine-, leucine-, and phenylalanine-based peptide monolayers revealed that hydrophobic and positively charged peptides decreased non-specific adsorption when using lysate from HEK 293FT cells. A polyethylene glycol (PEG) monolayer resulted in 2-fold greater fouling than the best peptide [3-MPA−(His)2(Leu)2(Phe)2−OH] under the same conditions. Matrix-assisted laser desorption ionization tandem time-of-flight mass spectrometry (MALDI−TOF/TOF MS) analysis of the adsorbate from cell lysate confirmed that lipids are the main source of non-specific adsorption. Importantly, the mass spectrometry (MS) study revealed that both the number of lipids identified and their intensity decreased with decreasing non-specific adsorption. A peptide monolayer thus provides an efficient mean to suppress non-specific adsorption from this human cell lysate.



INTRODUCTION The healthcare industry constantly searches for methodologies to improve diagnostic techniques, in terms of analysis time, sensitivity, and selectivity. Despite the numerous technological and medical advances, cancer remains one of the biggest scourges of humanity. The need is thus well-justified to develop early and reliable cancer diagnostic technologies to improve the chance of remission of cancer patients.1 Enzyme-linked immunosorbent assays (ELISAs) and various biosensing technologies are the staple methodologies for identifying and quantifying cancer biomarkers. These technologies rely mainly on solubilized proteins found in serum because of the noninvasive nature of sample collection. Nonetheless, not all proteins are of diagnostic interest. Many biosensors have been developed using a large variety of selective bioreceptors, such as enzymes, antibodies, or DNA.2 Different techniques, such as mass spectrometry (MS)3,4 or fluorescence spectroscopy employed for protein arrays,5 provide analytical detection of the results on the sensor. Surface plasmon resonance (SPR) is an analytical tool of particular interest because it allows real-time monitoring of © 2013 American Chemical Society

multiple biomarkers with sensitivity and rapidity. SPR sensing has been significantly developed in recent years to quantify biomarkers in complex matrixes, such as serum,6−8 plasma,9 or saliva.10 SPR biosensors have been shown to be suited to detection of several cancer biomarkers in solubilized samples, such as oropharyngeal squamous cell carcinoma,10 prostate cancer,6 colorectal, gastric, and pancreatic cancer,7 intestinal cancer,8 liver cancer,9 and ovarian cancer,11 among others. However, several types of cancer do not exhibit a significant protein or metabolomic response in biofluids, such as blood, serum, urine, or saliva. In those cases, biopsies are taken from the suspected tumor site to verify the presence of cancerous cells. The staple diagnostic method, immunostaining and microscopic readout, is lengthy, requires highly skilled pathologists, and suffers from subjective interpretation. This procedure is thus prone to human error, which may delay the assessment of the presence of cancer and retard the onset of Received: May 15, 2013 Revised: July 9, 2013 Published: July 11, 2013 10141

dx.doi.org/10.1021/la401837y | Langmuir 2013, 29, 10141−10148

Langmuir

Article

Microscope cover glass slides (22 × 22 mm, VWR) were cleaned for 90 min with a hot Piranha solution (Caution! Piranha solution is highly corrosive). The slides were then coated with a 1 nm Cr adhesion layer, followed by a 50 nm thick Au layer to create the SPR sensor. The sensors were reacted overnight in a 1 mM DMF solution of the peptide to form the monolayer. The sensors were then rinsed with ethanol to wash away the unbound peptides. Detailed studies of the formation and conformation of the peptide monolayers concluded that depositing thiolated peptides at open-circuit potential forms a densely packed and single monolayer on the surface of Au sensors.19−23 A full monolayer is necessary to suppress non-specific adsorption. Cell Lysate. HEK 293FT cells were cultured at 37 °C with a 5% CO2 atmosphere in a Dulbecco’s modified Eagle’s medium (DMEM) (−) cell growth medium (Life Technologies), in which 10% (v/v) fetal bovine serum (Life Technologies) and 1% (v/v) of a 100× penicillin/ streptomycin solution (Life Technologies) were added to obtain DMEM (+) culture media. Cells were passaged and propagated to obtain pools of 109 cells. The pooled cells were lysed with 50 mL of radioimmunoprecipitation assay (RIPA) buffer composed of N-2hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES) (50 mM), NaCl (250 mM), ethylenediaminetetraacetic acid (EDTA) (5 mM), Triton-X100 (1%, v/v), and a protease inhibitor cocktail tablet (Complete ULTRA tablets, glass vials from Roche Applied Science). The mixture was passed through a BD-25G11/2 needle to complete the lysis. Quantification of lipids in cell lysate was performed as described below. A common Bligh and Dyer extraction was used to extract lipids.24 Specifically, to 1 mL of cell lysate was added 3.75 mL of a 1:2 chloroform/methanol extraction solvent. After vortexing, 1.25 mL of chloroform was added, followed by the same volume of water after vortexing. The mixture was vortexed again and then allowed to separate into two phases. The chloroform phase was recovered with a Pasteur pipet, washed with 2.25 mL of water, and transferred in a preweighed vial. After the chloroform was evaporated, the vial was weighed again to quantify the lipids. SPR Measurements. The SPR chip was mounted on a custommade SPR instrument based on a dove prism25 and allowed to stabilize for 15 min. Optical contact was ensured with refractive index (RI)matching oil, deposited between the SPR chip and the prism (requires stabilization). Then, 100 spectra of s-polarized light were recorded as a reference. Each spectrum was acquired at 1 Hz from 10 accumulations at 100 ms integration time. p-Polarized light served for the real-time measurements. The peptide-modified SPR chips were exposed to phosphate-buffered saline (PBS) for 5 min and to cell lysate for 20 min, and the baseline was monitored again in PBS for 5 min. Data were processed with Matlab software. Each peptide SAM was subjected to this experiment at least 4 times to study reproducibility of the results. MALDI−Tandem Time-of-Flight (TOF/TOF) MS of Nonspecifically Bound Molecules. MALDI−TOF/TOF analyses were performed to monitor peptides and lipids on the adsorbed layer of cell lysate on peptide SAMs using α-cyano-4-hydroxycinnamic acid (CHCA, with peptides and lipids in positive mode) and 2′,6′dihydroxyacetophenone (DHA, with lipids in negative mode) matrixes. Matrix solutions were prepared at 10 mg/mL in 1:1 acetonitrile H2O + 0.1% trifluoroacetic acid. It is noteworthy to point out that the slides used for MS experiments were 25 × 75 × 1.0 mm microscope slides to fit the slide adaptor of the MS instrument. The slides were Au-coated and functionalized, as described above. Functionalized slides were incubated for 10 min in cell lysate and then rinsed for 2 min in PBS and 2 min in distilled water. A 30 min trypsin digestion was run at 37 °C on three different spots (3 μL) per slide on a heating plate controlled by a temperature sensor. The substrate was mounted on hydrated cotton wool on the heating plate to distribute heat uniformly. A reference gold-coated slide with a drop of water was used as a reference for the temperature sensor. Matrix deposition was performed by successive evaporation of 0.5 μL of matrix solution on the digested spots to analyze proteins and beside the digested spots for lipid analysis. MS analysis of the adsorbed material was performed with a MALDI−TOF/TOF mass spectrom-

treatment. To circumvent this problem, direct lysis of biopsy tissue for biomarker analysis with a SPR biosensor could provide analytical-grade results in less than 1 h. Cell lysate is a lipid-rich complex biological matrix, which is expected to significantly foul the sensor surface. Although lipid adsorption promoted by different surface chemistries has been used to immobilize active membrane proteins on sensors12 and as a membrane-cloaking technique to analyze proteins in serum,13 SPR sensing requires low fouling surface chemistries. A recent review details the significant efforts deployed to circumvent the non-specific adsorption problem in biosensing.14 The most promising surface chemistries include a polycarboxybetaine acrylamide surface15 and polyMeOEGMA16 surfaces. Recently, a peptidoid polymer17 and a peptide18 selfassembled monolayer (SAM) were proposed to reduce nonspecific adsorption of serum. All of these surfaces have been developed to prevent non-specific adsorption of proteins from serum, but to our knowledge, little information is available about low-fouling surface chemistry for cell lysate. As a result of the lipid-rich nature of cell lysate in opposition to the proteinrich composition of serum, it is expected that the surfaces developed to prevent non-specific adsorption from serum will not be effective when used with cell lysate. We have previously applied a surface developed for serum analysis toward analysis using crude bacterial cell lysate, with moderate success.19 Although the enzyme β-lactamase was successfully detected in the test cell lysate, significant fouling was observed on the sensor exposed to a control cell lysate. The need thus exists to develop a surface chemistry suited to reduce the non-specific adsorption from cell lysate. Establishing the physical parameters governing non-specific interactions from cell lysate is primordial to design low fouling surface chemistries. Understanding non-specific adsorption implies obtaining chemical information on the complex processes involved at surfaces. SPR sensing provides quantitative information on the total “mass” adsorbed to the chip. To provide information on the chemical identity of molecular adsorbates, a combination of SPR and MS must be used.20 To date, SPR−MS has been mainly applied to identify protein−antibody interactions.21 While this technique has tremendous potential, it has never been employed for identification of non-specifically bound molecules. We show in this paper a first example that makes use of matrix-assisted laser desorption ionization (MALDI) time-of-flight (TOF) MS for identification of lipids and proteins non-specifically bound to the SPR surface.



EXPERIMENTAL SECTION

Peptide-Modified SPR Chips. A solid-phase procedure was extensively described in the literature to synthesize the 3-MPA− peptides−OH18,19 and was adapted as detailed below. After the first coupling, the resin was protected by reacting 10 equiv of Nethyldiisopropylamine (Alfa Aesar) and 10 equiv of acetic anhydride (Sigma Aldrich) for 1 h. Also, only 6 equiv of N-ethyldiisopropylamine instead of 9 equiv were used for each amino acid coupling. Finally, the resin was rinsed only 3 times with N,N-dimethylformamide (DMF) between each step, except when the resin had to be stored dry overnight, in which case the DMF−methanol−dichloromethane (DCM) rinsing described in the literature was performed. The exact mass of each peptide was verified using MS (ESI−TOF−MS, Agilent), with a direct injection of a 1:1 (w/v) solution of the peptide in ethanol. Because of the electrospray ionization (ESI) process, singly and multiply charged ions attributed to the expected peptides were observed. 10142

dx.doi.org/10.1021/la401837y | Langmuir 2013, 29, 10141−10148

Langmuir

Article

eter (UltrafleXtreme, Bruker Daltonics, Billerica, MA) as previously described by Ratel et al.26 Protein and lipid identifications were made with MASCOT search engine and LIPID MAPS database (http:// www.lipidmaps.org), respectively.

Table 1. Influence of the Hydrophobicity of the Amino Acid on Non-specific Adsorption from Cell Lysate on Pentapeptide SAMs



RESULTS AND DISCUSSION Non-specific Adsorption of Homopeptides. It has been long established that non-specific adsorption of proteins in serum is reduced by hydrophilic surfaces,27 such as hydrophilic peptide SAMs.19−22 It is however essential to reassess the impact of the surface property in the face of the distinctive molecular composition of cell lysate.29 Here, we focus on lysate from cultured human cells in the interest of improving diagnostic capacity using lysate from biopsies. To monitor the effect of peptide composition on non-specific adsorption, 3MPA−(AA)5−OH SAMs of peptides with differing polarities were synthesized (Figure 1). Hydrophobic and hydrophilic

peptide 3-MPA−(Ser)5−OH 3-MPA−(Asp)5−OH 3-MPA−(His)5−OH 3-MPA−(Phe)5−OH 3-MPA−(Leu)5−OH

amino acids (serine, aspartic acid, histidine, leucine, and phenylalanine) were tested to cover the effects of charge, polarity, and aromaticity. Non-specific surface coverage after the exposition of the SAMs to HEK 293FT cell lysate was measured from the SPR response [in wavelength shift (ΔλSPR)] converted to surface coverage (Γ) by the equation derived from the equation proposed by Jung et al.30 ⎞ −ld ⎛ ΔλSPR ⎟⎟ ln⎜⎜1 − 2 m(ηads − ηsol ) ⎠ ⎝

%bilayer adsorbed

± ± ± ± ±

92 91 55 47 39

458 454 274 235 194

97 162 58 77 77

contact anglePBS (deg) (ref 19) 21 14 26 74 75

± ± ± ± ±

2 2 1 2 1

the SPR chip. Interestingly, the positive polyhistidine peptide was relatively efficient at reducing non-specific adsorption of cell lysate (274 ng/cm2). Leucine- and phenylalanine-based peptides were significantly more effective with respect to cell lysate adsorption (nearly 200 ng/cm2 or 40% of a bilayer), although these peptide monolayers performed poorly with serum. Contact angles were previously measured in PBS for this set of peptides.19 The most hydrophilic peptides, as measured by contact angle, were the most susceptible to be fouled with cell lysate. This heavily contrasts with the previous conclusions for serum, stating that hydrophilic surfaces better resist nonspecific adsorption from serum.27−29 It is interesting to observe that different physicochemical properties reduce non-specific adsorption from cell lysate. The chain length of the peptide monolayers significantly influenced the extent of non-specific adsorption from serum.19 Three amino acids were selected to evaluate the effect of the chain length on non-specific adsorption of cell lysate. A neutral and polar (serine), a hydrophobic (leucine). and positively charged (histidine) amino acid were used to build 3-MPA− (AA)n−OH peptides, where n = 1−5. The chain length had little influence on the non-specific adsorption process (Table 2). Thus, the hydrophobicity trend observed for pentapeptides

Figure 1. General structure of 3-MPA−amino acid−OH-type peptides.

Γ=ρ

Γ (ng cm2)

Table 2. Non-specific Adsorption from Diluted Cell Lysate on Peptide SAMs of Different Lengths

(1)

peptide 3-MPA−(Ser)1−OH 3-MPA−(Ser)2−OH 3-MPA−(Ser)3−OH 3-MPA−(Ser)4−OH 3-MPA−(Ser)5−OH 3-MPA−(His)1−OH 3-MPA−(His)2−OH 3-MPA−(His)3−OH 3-MPA−(His)4−OH 3-MPA−(His)5−OH 3-MPA−(Leu)1−OH 3-MPA−(Leu)2−OH 3-MPA−(Leu)3−OH 3-MPA−(Leu)4−OH 3-MPA−(Leu)5−OH

where the density of the monolayer (ρ) is 1.05 g/cm3,31 the penetration depth of the surface plasmons28 (ld) is approximately 230 nm, and the refractive index of the adsorbed material (ηads) is approximately 1.5 refractive index units (RIU).32 The sensitivity of the biosensor was previously established as 1765 ± 100 nm/RIU.28 Finally, the refractive index of cell lysate (ηsol) was measured with a high-resolution refractometer at 1.339 32 ± 0.000 02 RIU. We estimate the surface coverage of a full lipid bilayer to be 0.5 μg/cm2 of adsorbed materials using an average value of 750 g/mol for lipids and 0.5 nm2/lipid molecule in the lipid bilayer.28 The fraction of a lipid bilayer resulting from non-specific adsorption can be estimated from these values. The comparison of peptides constituted with different classes of amino acids reveals that the peptides built with hydrophobic amino acids are more efficient in preventing cell lysate adsorption on the sensor surface than hydrophilic peptides (Table 1). Serine- and aspartic-acid-based peptides, shown to be low fouling with respect to serum, showed significant fouling with cell lysate, with nearly 450 ng/cm2 adsorbed to the surface. Assuming that adsorption was essentially due to lipids (confirmed by MS analysis; see below), this corresponds to a lipid bilayer surface coverage over nearly 90% of the surface of

Γ (ng cm2)

%bilayer adsorbed

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

51 53 119 72 92 59 48 54 48 55 37 29 45 46 39

256 265 597 361 458 296 241 269 241 274 183 147 223 230 194

47 101 152 211 97 184 86 76 63 58 54 44 108 148 77

contact anglePBS (deg) (ref 19) 38 37 25 22 21 32 29 27 27 26 63 70 71 74 75

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

4 1 2 1 2 2 3 2 1 1 1 2 1 1 1

remained unchanged when the chain length was varied, with hydrophobic peptides being generally more effective at preventing fouling from cell lysate than hydrophilic peptides. Contact angles were previously measured for these sets of peptides, and increasing hydrophobicity followed the series serine < histidine < leucine (Table 2 and ref 19), inversely proportional to the non-specific adsorption observed with cell 10143

dx.doi.org/10.1021/la401837y | Langmuir 2013, 29, 10141−10148

Langmuir

Article

Table 3. Non-specific Adsorption from Diluted Cell Lysate on Diblock Pentapeptide SAMs peptide 3-MPA−(Phe)1(Leu)4−OH 3-MPA−(Phe)2(Leu)3−OH 3-MPA−(Phe)3(Leu)2−OH 3-MPA−(Phe)4(Leu)1−OH 3-MPA−(Phe)5−OH 3-MPA−(Leu)1(Phe)4−OH 3-MPA−(Leu)2(Phe)3−OH 3-MPA−(Leu)3(Phe)2−OH 3-MPA−(Leu)4(Phe)1−OH 3-MPA−(Leu)5−OH

Γ (ng cm2)

%bilayer adsorbed

± ± ± ± ± ± ± ± ± ±

42 40 45 44 47 30 33 22 35 39

210 201 225 220 235 148 165 112 173 194

84 35 57 79 77 54 87 28 52 77

contact anglewater (deg) 57 41 61 59 67 64 66 73 60 75

± ± ± ± ± ± ± ± ± ±

2 3 2 1 3 2 1 2 1 1

demonstrates that the peptide sequence has a critical impact on non-specific adsorption. Water contact angles were measured for the series of peptides with leucine and phenylalanine on the C terminus. Generally, a C-terminal phenylalanine increased the water contact angle and exhibited lower non-specific adsorption than the more hydrophilic peptides with a C-terminal leucine. The solvent-exposed C-terminal position plays a major role in reducing non-specific adsorption. Thus, the simple inversion of two blocks in the sequence modified the surface and nonspecific adsorption properties of the SAM. 3-MPA− (Leu)3(Phe)2−OH exemplifies this concept, offering the most hydrophobic surface and resulting in the lowest non-specific adsorption among the diblock peptides, while its inverted counterpart 3-MPA−(Phe)2 (Leu)3−OH shows a nearly 2-fold lower contact angle and results in nearly 2-fold higher nonspecific adsorption. Various triblock peptides having the overall structure 3MPA−AABBCC−OH, 3-MPA−(ABC)2−OH, or 3-MPA− (AB)3−OH were synthesized to further improve on the conclusions drawn from homopeptides (Tables 1 and 2) and diblock peptides (Table 3), where A, B, and C are either leucine, phenylalanine, or histidine. The non-specific adsorption on triblock peptides was measured with a more highly concentrated cell lysate, providing a more stringent test for the triblock peptides and the reference materials [polyethylene glycol (PEG) and Au]. Because many factors influence nonspecific adsorption, data can be directly compared only if they are obtained from a similar quality of cell lysate. To allow for a comparison between sets of experiments, the non-specific adsorption experiments were repeated for some key peptides from Tables 1−3 using the more concentrated cell lysate (Table 4). Non-specific adsorption to 3-MPA−(Ser)5−OH and 3-MPA−(Leu)3(Phe)2−OH increased by approximately a factor of 3−6 with the more concentrated cell lysate. Further, to ensure that those differences in surface coverage were attributed to the concentration of cell lysate, a Bligh and Dyer

lysate. The chain length significantly influences the properties of peptide SAM.19 The surface coverage of the monolayer (thus, the conformation), the non-specific adsorption of serum, and the contact angle all depend upon the chain length of short-chain peptides (n = 1−5). We note that the serine series shows the largest changes in the contact angle. The shortest chains of the serine series are the most hydrophobic peptides and also the serine peptides with the lowest non-specific adsorption. The changes in hydrophobicity were significantly smaller for the other series, and thus, the changes in nonspecific adsorption were also marginal. Thus, hydrophobicity is a major physical chemistry parameter influencing non-specific adsorption, but in the case of cell lysate, more hydrophobic surfaces are preventing non-specific adsorption. This conclusion contrasts with the general trend of hydrophilic surfaces being better at preventing non-specific adsorption of serum or protein-based biofluids. It is expected that an entirely different set of surface chemistry properties would be required to further reduce nonspecific adsorption from cell lysate. The main component of serum is protein, while lipids are the major constituents of cell lysate. Lipids in cell lysate are essentially a concentrated solution of fragments of bilayers, the surface of which is quite hydrophilic as a result of the charged polar headgroups of the lipids. On the basis of the observations by Kyo et al.33 that negative charges increase non-specific adsorption of cell lysate, we could easily hypothesize that histidine-based peptides would perform better than other hydrophilic or hydrophobic peptides because they bear basic groups, which could electrostatically repel the lipid bilayers and reduce non-specific adsorption. However, we conclude that the hydrophobic character of leucine-based peptide contributes significantly in reducing nonspecific adsorption lipid bilayer fragments contained in HEK 293FT cell lysate. Thus, both positively charged and hydrophobic side chains are efficient at suppressing non-specific adsorption of cell lysate. Non-specific Adsorption on Di- and Triblock Peptides. Diblock peptides were previously shown to reduce non-specific adsorption in the presence of crude serum.22 Thus, diblock peptides 3-MPA−AnBm−OH were synthesized for A and B being either leucine or phenylalanine; n and m were varied from 1 to 4, and n + m was equal to 5 for every peptide. The resulting peptides terminated with phenylalanine were found to have a better capacity to reduce non-specific adsorption from cell lysate down to nearly 150 ng/cm2 or 30% of a bilayer (Table 3). The leucine-terminated peptides showed constant non-specific adsorption of nearly 215 ng/cm2, corresponding to 43% of a bilayer. Non-specific adsorption was decreased to 112 ng/cm2 with 3-MPA−(Leu)3(Phe)2−OH. This analysis clearly

Table 4. Non-specific Adsorption from Concentrated Cell Lysate on Triblock Hexapeptide SAMs peptide 3-MPA−(His)2(Phe)2(Leu)2−OH 3-MPA−(Leu)2(His)2(Phe)2−OH 3-MPA−((His)(Leu)(Phe))2−OH 3-MPA−((Phe)(His)(Leu))2−OH 3-MPA−((Leu)(Phe))3−OH 3-MPA−((Leu)(Phe)(His))2−OH 3-MPA−(Phe)2(Leu)2(His)2−OH 3-MPA−(His)2(Leu)2(Phe)2−OH 10144

Γ (ng cm2)

%bilayer adsorbed

± ± ± ± ± ± ± ±

155 153 149 148 118 111 37 32

774 763 744 739 592 554 187 159

57 102 92 95 126 22 10 27

dx.doi.org/10.1021/la401837y | Langmuir 2013, 29, 10141−10148

Langmuir

Article

specific adsorption as the hydrophobic peptide developed here. PEG and 3-MPA−[(Leu)(His)(Asp)]2−OH were 2- and 5fold, respectively, worse than 3-MPA−(His)2(Leu)2(Phe)2− OH (Table 5). Again, the trend of more hydrophobic surfaces was confirmed with contact angle measurements for peptide monolayers. While the hydrophobicity and charge definitely play a major role in driving non-specific adsorption processes on the SPR sensors, they are not exclusive. Bare gold is very hydrophobic but fouls significantly, while PEG is quite hydrophilic and negatively charged at the pH of the buffer and performs moderately well. Hydrophobic and/or positively charged peptides definitely improve non-specific adsorption of cell lysate and, thus, constitute a general rule of thumb for minimizing non-specific adsorption. Our conclusions are in agreement to the observations by Kyo et al.33 that negatively charged monolayers tend to favorably adsorb cell lysate. They observed that a full monolayer of PEG−COOH was adsorbing a significant amount of cell lysate and suppressed non-specific adsorption by diluting PEG−COOH in a mixed monolayer. We confirmed that PEG−COOH does absorb cell lysate and that peptide monolayers can further decrease non-specific adsorption. A further comparison is difficult because the sources and type of cell lysate differ between studies. MS Analysis of Adsorbed Molecules from Cell Lysate. While the non-specific adsorption study points to lipid adsorption, cell lysate is a complex matrix containing a diversity of proteins and small molecules in addition to lipids. MS analysis provides molecular fingerprinting of the materials nonspecifically adsorbed on various surface chemistries. Protein adsorption to surfaces is well-documented for serum-based analysis.14−34 Because non-specific adsorption from cell lysate has never been analyzed by MS, the interaction between the peptide SAMs and cell lysate must be investigated. Nonspecifically adsorbed proteins were analyzed after SAM trypsin digestion of the proteins. MS analyses of the resulting tryptic peptides were performed by MALDI MS and tandem mass spectrometry (MS/MS), and identification of the corresponding proteins was performed by database interrogation with the MASCOT search engine. Identification of the proteins adsorbed to the different SAM peptide surfaces is presented in Table 6. Results from the protein analysis are of particular interest for two reasons. The proteins identified, such as collagen α-1(XX) chain, 40S ribosomal proteins (S18, S19, and S29), and mitogen-activated protein (kinase)55 are known to be present in HEK 293FT cell lysate. Surprisingly, these proteins were detected on only one [3-MPA−(Leu)5−OH] of the seven peptide SAMs. None of the other surfaces tested [3-MPA− (Ser)5−OH, 3-MPA−(Phe)5−OH, 3-MPA−(Leu)3(Phe)2− OH, 3-MPA−(Phe)1(Leu)4−OH, 3-MPA−(His)2(Leu)2(Phe)2−OH, and 3-MPA−(Phe)2(Leu)2(His)2− OH] held any identifiable tryptic peptide. The MS analysis

lipid extraction was performed on 1 mL of cell lysate. The mass of lipids obtained after drying indicated that the second cell lysate was 3 times more concentrated than the former (6 mg/ mL compared to 2 mg/mL), which correlates with the nonspecific adsorption results. As previously demonstrated for serum,18 triblock peptides significantly reduced non-specific adsorption from cell lysate. In particular, 3-MPA−(His)2(Leu)2(Phe)2−OH reduced the nonspecific adsorption from the concentrated cell lysate to a surface coverage of 159 ng/cm2, corresponding to 31% of a lipid bilayer. In comparison, the best diblock peptide 3-MPA− (Leu)3(Phe)2−OH fared significantly worse, with 698 ng/cm2 or 136% of a lipid bilayer from the concentrated cell lysate. The triblock peptide is thus more efficient in terms of reducing nonspecific adsorption. The optimal peptide provides 6 times lower non-specific adsorption from cell lysate than unprotected bare gold. The peptides with a C-terminal phenylalanine or histidine performed better than those with a C-terminal leucine. This strongly suggests that a combination of positively charged surfaces and hydrophobicity is required to minimize nonspecific interactions. In addition, the position of the blocks within each peptide clearly have an important impact on performance, as shown by comparison of 3-MPA−((Leu)(Phe)(His))2−OH (554 ng/cm2) and 3-MPA− (Phe)2(Leu)2(His)2−OH (187 ng/cm2). To confirm the incompatibility of surfaces designed for use in serum with respect to use with cell lysate, non-specific adsorption measurements were conducted using either a thiolated PEG [HS−(CH2)11−(EG)4−OCH2−COOH] SAM, which is commonly used to prevent serum adsorption on biosensors, or a 3-MPA−[(Leu)(His)(Asp)]2−OH SAM (Figure 2), because it has been found to be the most efficient

Figure 2. Overview of reduction of non-specific adsorption from cell lysate with peptide SAMs: 3-MPA−(Ser)5−OH (dark blue), Au (red), 3-MPA−(Leu)3(Phe)2−OH (green), HS−PEG (light blue), and 3MPA−(His)2(Leu)2(Phe)2−OH (pink).

peptide for reducing non-specific adsorption from serum.14 Neither of these SAMs offered as great of a reduction in non-

Table 5. Comparison of Non-specific Adsorption from Concentrated Cell Lysate on Various Surfaces peptide 3-MPA−(Ser)5−OH Au 3-MPA−[(Leu)(His)(Asp)]2−OH 3-MPA−(Leu)3(Phe)2−OH PEG 3-MPA−(His)2(Leu)2(Phe)2−OH

Γ (ng cm2)

%bilayer adsorbed

± ± ± ± ± ±

231 186 158 83 59 32

1156 929 789 413 294 159

141 186 107 47 25 27 10145

contact anglewater (deg) 45 87 53 71 25 77

± ± ± ± ± ±

3 2 1 3 3 4

dx.doi.org/10.1021/la401837y | Langmuir 2013, 29, 10141−10148

Langmuir

Article

efficient peptide SAM for preventing non-specific adsorption of cell lysate results in a smaller variety of lipid signals found on the surface, which is in strong agreement with the SPR measurements. For example, 3-MPA−(Ser)5−OH exhibited the greatest number of lipids identified, while 3-MPA− (Phe)2(Leu)2(His)2−OH had the fewest number of lipids identified (Table 7). We observed that most of the lipids identified were in the phosphatidyl choline family, which are quite hydrophilic and have a positively charged headgroup. We note that several of the surfaces analyzed [3-MPA−(Leu)5− OH, 3-MPA−(Leu)2(Phe)3−OH, 3-MPA−(Phe)1(Leu)4−OH, and 3-MPA−(His)2(Leu)2(Phe)2−OH] did not give rise to mass signals that could be attributed to lipids (Table 7). These results confirm the capacity of hydrophobic and/or positively charged peptide SAMs to minimize lipid adsorption.

Table 6. MALDI−TOF/TOF Analysis of Non-specifically Adsorbed Protein on 3-MPA−(Leu)5−OH SAMa m/z 801.363 1134.551 1247.579 1338.808 1408.664 1703.794

ion [M [M [M [M [M [M

+ + + + + +

H]+ H]+ H]+ H]+ H]+ H]+

protein collagen α-1(XX) chain 40s ribosomal protein S19 40s ribosomal protein S18 mitogen-activated protein (kinase)55 40s ribosomal protein S29 40s ribosomal protein S19

a This analysis has also been made on 3-MPA−(Ser)5−OH, 3-MPA− (Phe)5−OH, 3-MPA−(Leu)3(Phe)2−OH, 3-MPA−(Phe)1(Leu)4− OH, 3-MPA−(His)2(Leu)2(Phe)2−OH, and 3-MPA− (Phe)2(Leu)2(His)2−OH SAMs, but no protein could be identified because there was simply no peak that could be linked to a protein on the MS spectrum.



CONCLUSION This paper reports the development of a novel surface chemistry that will improve SPR analysis in crude cell lysate. Even though a similar experiment has previously been reported for other complex matrixes, such as serum, the results obtained in serum were shown here to be incompatible with the use of cell lysate. PEG and our best-performing peptide for serum analyses, 3-MPA−[(Leu)(His)(Asp)]2−OH, were significantly worse than 3-MPA−(His)2(Leu)2(Phe)2−OH at preventing surface fouling. Peptide monolayers with hydrophobic and basic groups showed greater potential to reduce significantly nonspecific adsorption of cell lysate. In addition, the nature of the

thus confirms that protein adsorption from cell lysate is not an important source of fouling. Because lipids are the major component of cell lysate, a thorough MS analysis for lipids was performed by MALDI MS with both positive and negative polarities. A first MS analysis was performed to obtain the general lipid spectrum for each SAM. Then, every peak which mass-to-charge value could be attributed to a lipid was analyzed by MS/MS for identification (Figure 3). Hence, with the molecular ion and the MS/MS spectra, some lipids could be identified. Table 7 reports the most likely lipids observed considering the mass spectral accuracy and the assignment power of the software. The most

Figure 3. Mass spectra of adsorbed lipids on 3-MPA−(Ser)5−OH SAM with CHCA matrix: (A) zoom on the lipid mass range and (B) MS/MS spectra of the 761 Da lipid. 10146

dx.doi.org/10.1021/la401837y | Langmuir 2013, 29, 10141−10148

Langmuir

Article

Table 7. MALDI−TOF/TOF Analysis of Non-specifically Adsorbed Lipids on Various SAMsa SAM m/z

ion

lipid

636.5 650.5

[M + H]+ [M + H]+

PC(O-16:0/10:0) PC(12:0/14:0)

666.5 676.5 678.5 694.9 706.6 732.6 734.6 760.6

[M [M [M [M [M [M [M [M

+ + + + + + + +

H]+ H]+ H]+ Na]+ H]+ H]+ H]+ H]+

SM(d18:0/12:0) PC(28:1)* PC(28:0)* PC(28:3)* PC(30:0)* PC(32:1)* PC(14:0/18:0) PC(16:0/18:1)

786.6 831.2 857.3 849.6

[M [M [M [M

+ H]+ + K]+ + K]+ − H]−

PC(18:0/18:2) PA(22:5/22:6) PI(O-16:0/18:3) PG(20:5/22:1)

901.4 941.6 1013.8

[M − H]− [M − H]− [M − H]−

PI(20:5/20:5) PI(22:4/20:0) PIP(20:3/22:5)

MS/MS peaks

3-MPA−(Ser)5−OH

3-MPA−(Phe)5−OH

3-MPA−(Leu)3(Phe)2−OH

× ×

×

×

184, 395 184, 422, 439 184, 467 184 184 184 184 184 184, 524 184, 505, 576 184, 503 468 278 301, 337, 529 301, 600 311, 315 305, 313, 329, 338

3-MPA−(Phe)2(Leu)2(His)2−OH

× × × × × × × × × × × × × × ×

a This analysis was repeated on 3-MPA−(Leu)5−OH, 3-MPA−(Leu)2(Phe)3−OH, 3-MPA−(Phe)1(Leu)4−OH, and 3-MPA−(His)2(Leu)2(Phe)2− OH SAMs, but no lipid could be identified, because either no lipid given by LIPID MAPS could be attributed to the cellular lysate used or there was simply no peak that could be linked to a lipid on the MS spectrum.

(3) Wei, B.-R.; Hoover, S. B.; Ross, M. M.; Zhou, W.; Meani, F.; Edwards, J. B.; Spehalski, E. I.; Risinger, J. I.; Alvord, W. G.; Quinones, O. A.; Belluco, C.; Martella, L.; Campagnutta, E.; Ravaggi, A.; Dai, R.M.; Goldsmith, P. K.; Woolard, K. D.; Pecorelli, S.; Liotta, L. A.; Petricoin, E. F., III; Simpson, R. M. Serum S100A6 concentration predicts peritoneal tumor burden in mice with epithelial ovarian cancer and is associated with advanced stage in patients. PLoS One 2009, 4 (10), No. e7670. (4) Sun, Y.; Zang, Z.; Xu, X.; Zhang, Z.; Zhong, L.; Zan, W.; Zhao, Y.; Sun, L. Differential proteomics identification of HSP90 as potential serum biomarker in hepatocellular carcinoma by two-dimensional electrophoresis and mass spectrometry. Int. J. Mol. Sci. 2010, 11 (4), 1423−1433. (5) Ramirez, A. B.; Loch, C. M.; Zhang, Y. Z.; Liu, Y.; Wang, X. H.; Wayner, E. A.; Sargent, J. E.; Sibani, S.; Hainsworth, E.; Mendoza, E. A.; Eugene, R.; LaBaer, J.; Urban, N. D.; McIntosh, M. W.; Lampe, P. D. Use of a single-chain antibody library for ovarian cancer biomarker discovery. Mol. Cell. Proteomics 2010, 9 (7), 1449−1460. (6) Choi, S.; Chae, J. Methods of reducing non-specific adsorption in microfluidic biosensors. J. Micromech. Microeng. 2010, 20 (7), 075015. (7) Su, F.; Xu, C.; Taya, M.; Murayama, K.; Shinohara, Y.; Nishimura, S.-I. Detection of carcinoembryonic antigens using a surface plasmon resonance biosensor. Sensors 2008, 8 (7), 4282−4295. (8) Ladd, J.; Lu, H.; Taylor, A. D.; Goodell, V.; Disis, M. L.; Jiang, S. Direct detection of carcinoembryonic antigen autoantibodies in clinical human serum samples using a surface plasmon resonance sensor. Colloids Surf., B 2009, 70 (1), 1−6. (9) Teramura, Y.; Iwata, H. Label-free immunosensing for αfetoprotein in human plasma using surface plasmon resonance. Anal. Biochem. 2007, 365 (2), 201−207. (10) Yang, C. Y.; Brooks, E.; Li, Y.; Denny, P.; Ho, C. M.; Qi, F. X.; Shi, W. Y.; Wolinsky, L.; Wu, B.; Wong, D. T. W.; Montemagno, C. D. Detection of picomolar levels of interleukin-8 in human saliva by SPR. Lab Chip 2005, 5 (10), 1017−1023. (11) Liu, C.; Lei, T.; Ino, K.; Matsue, T.; Tao, N.; Li, C.-Z. Real-time monitoring biomarker expression of carcinoma cells by surface plasmon resonance biosensors. Chem. Commun. 2012, 48 (84), 10389−10391.

material adsorbed on the SAMs was confirmed by MALDI− TOF/TOF MS, which revealed that few proteins adsorbed on the surface, while lipids constituted the major source of nonspecific adsorption.



AUTHOR INFORMATION

Corresponding Author

*Telephone: +1-514-343-7342. E-mail: jf.masson@umontreal. ca. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS Financial support was provided by the Natural Sciences and Engineering Research Council of Canada (NSERC), the Center for Self-Assembled Chemical Structures (CSACS), the Fond québécois de la recherche sur la nature et les technologies (FQRNT), the Canadian Foundation for Innovation (CFI), the Mérieux Research Grants, Université de Montréal, and Univalor. We thank Stéphane Roy for the HEK 293FT cell line and use of cell culture facilities and Damien Colin for producing cell lysate. The authors also acknowledge the help of Aurélien Thomas (Département de Chimie, Université de Montréal) for his help with searches in LIPID MAPS.



REFERENCES

(1) Reddy, P. J.; Sadhu, S.; Ray, S.; Srivastava, S. Cancer biomarker detection by surface plasmon resonance biosensors. Clin. Lab. Med. 2012, 32 (1), 47−72. (2) Shankaran, D. R.; Gobi, K. V. A.; Miura, N. Recent advancements in surface plasmon resonance immunosensors for detection of small molecules of biomedical, food and environmental interest. Sens. Actuators, B 2007, 121 (1), 158−177. 10147

dx.doi.org/10.1021/la401837y | Langmuir 2013, 29, 10141−10148

Langmuir

Article

lipid bilayer formation kinetics. Anal. Chem. 2008, 80 (13), 3666− 3676. (32) Giess, F.; Friedrich, M. G.; Heberle, J.; Naumann, R. L.; Knoll, W. The protein-tethered lipid bilayer: A novel mimic of the biological membrane. Biophys. J. 2004, 87 (5), 3213−3220. (33) Kyo, M.; Usui-Aoki, K.; Koga, H. Label-free detection of proteins in crude cell lysate with antibody arrays by a surface plasmon resonance imaging technique. Anal. Chem. 2005, 77 (22), 7115−7121. (34) Green, R. J.; Frazier, R. A.; Shakesheff, K. M.; Davies, M. C.; Roberts, C. J.; Tendler, S. J. B. Surface plasmon resonance analysis of dynamic biological interactions with biomaterials. Biomaterials 2000, 21 (18), 1823−1835.

(12) Besenicar, M.; Macek, P.; Lakey, J. H.; Anderluh, G. Surface plasmon resonance in protein−membrane interactions. Chem. Phys. Lipids 2006, 141 (1−2), 169−178. (13) Phillips, K. S.; Han, J. H.; Cheng, Q. Development of a “membrane cloaking” method for amperometric enzyme immunoassay and surface plasmon resonance analysis of proteins in serum samples. Anal. Chem. 2007, 79 (3), 899−907. (14) Blaszykowski, C.; Sheikh, S.; Thompson, M. Surface chemistry to minimize fouling from blood-based fluids. Chem. Soc. Rev. 2012, 41 (17), 5599−5612. (15) Brault, N. D.; Sundaram, H. S.; Li, Y.; Huang, C.-J.; Yu, Q.; Jiang, S. Dry film refractive index as an important parameter for ultralow fouling surface coatings. Biomacromolecules 2012, 13 (3), 589− 593. (16) Rodriguez-Emmenegger, C.; Kylian, O.; Houska, M.; Brynda, E.; Artemenko, A.; Kousal, J.; Bologna Alles, A.; Biederman, H. Substrateindependent approach for the generation of functional protein resistant surfaces. Biomacromolecules 2011, 12 (4), 1058−1066. (17) Lin, S.; Zhang, B.; Skoumal, M. J.; Ramunno, B.; Li, X.; Wesdemiotis, C.; Liu, L.; Jia, L. Antifouling Poly(β-peptoid)s. Biomacromolecules 2011, 12 (7), 2573−2582. (18) Bolduc, O. R.; Lambert-Lanteigne, P.; Colin, D. Y.; Zhao, S. S.; Proulx, C.; Boeglin, D.; Lubell, W. D.; Pelletier, J. N.; Fethiere, J.; Ong, H.; Masson, J.-F. Modified peptide monolayer binding His-tagged biomolecules for small ligand screening with SPR biosensors. Analyst 2011, 136 (15), 3142−3148. (19) Bolduc, O. R.; Clouthier, C. M.; Pelletier, J. N.; Masson, J. F. Peptide self-assembled monolayers for label-free and unamplified surface plasmon resonance biosensing in crude cell lysate. Anal. Chem. 2009, 81 (16), 6779−6788. (20) Nelson, R. W.; Krone, J. R.; Jansson, O. Surface plasmon resonance biomolecular interaction analysis mass spectrometry. 1. Chip-based analysis. Anal. Chem. 1997, 69 (21), 4363−4368. (21) Nedelkov, D. Development of surface plasmon resonance mass spectrometry array platform. Anal. Chem. 2007, 79 (15), 5987−5990. (22) Bolduc, O. R.; Pelletier, J. N.; Masson, J.-F. SPR biosensing in crude serum using ultralow fouling binary patterned peptide SAM. Anal. Chem. 2010, 82 (9), 3699−3706. (23) Bolduc, O. R.; Correia-Ledo, D.; Masson, J.-F. Electroformation of peptide self-assembled monolayers on gold. Langmuir 2012, 28 (1), 22−26. (24) Bligh, E. B.; Dyer, W. J. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 1959, 37 (8), 911−917. (25) Bolduc, O. R.; Live, L. S.; Masson, J.-F. High-resolution surface plasmon resonance sensors based on a dove prism. Talanta 2009, 77 (5), 1680−1687. (26) Ratel, M.; Branca, M.; Breault-Turcot, J.; Zhao, S. S.; Chaurand, P.; Schmitzer, A. R.; Masson, J.-F. Properties of ionic liquids on Au surfaces: Non-conventional anion exchange reactions with carbonate. Chem. Commun. 2011, 47 (38), 10644−10646. (27) Ostuni, E.; Chapman, R. G.; Holmlin, R. E.; Takayama, S.; Whitesides, G. M. A survey of structure−property relationships of surfaces that resist the adsorption of protein. Langmuir 2001, 17 (18), 5605−5620. (28) Alberts, B.; Johnson, A.; Lewis, J.; Raff, M.; Roberts, K.; Walter, P. Molecular Biology of the Cell, 5th ed.; Taylor and Francis Group: Oxford, U.K., 2007. (29) Ostuni, E.; Chapman, R. G.; Liang, M. N.; Meluleni, G.; Pier, G.; Ingber, D. E.; Whitesides, G. M. Self-assembled monolayers that resist the adsorption of proteins and the adhesion of bacterial and mammalian cells. Langmuir 2001, 17 (20), 6336−6343. (30) Jung, L. S.; Campbell, C. T.; Chinowsky, T. M.; Mar, M. N.; Yee, S. S. Quantitative interpretation of the response of surface plasmon resonance sensors to adsorbed films. Langmuir 1998, 14 (19), 5636−5648. (31) Mashaghi, A.; Swann, M.; Popplewell, J.; Textor, M.; Reimhult, E. Optical anisotropy of supported lipid structures probed by waveguide spectroscopy and its application to study of supported 10148

dx.doi.org/10.1021/la401837y | Langmuir 2013, 29, 10141−10148